8336( 5) - 04/05/00 09:33 - [ 1] PROPOSAL FOR HUBBLE SPACE TELESCOPE OBSERVATIONS ST ScI Use Only ID: 8336 Version: 5 Check-in Date: 05-Apr-2000 09:32:45 1.Proposal Title: A Unique Measurement of the Dust-to-Gas Ratio in an Absorption Line System ------------------------------------------------------------------------------------ 2. Proposal For 3. Cycle GO 8 ------------------------------------------------------------------------------------ 4. Investigators Contact? PI: Ross Cohen University of California - San Diego CoI: Dr. Edward Beaver University of California, San Diego N CoI: Dr. Margaret Burbidge University of California, San Diego N CoI: Dr. Vesa Junkkarinen University of California, San Diego N CoI: Mr. Ronald Lyons University of California, San Diego N ------------------------------------------------------------------------------------ 5. Abstract The dust--to--gas ratio in damped Ly-alpha absorbers has been determined from ISM metal abundances and from the effect of extinction on the background quasar continuua. Both methods have significant uncertainties. A damped system lies along the line-of-sight to both bright components of the gravitationally lensed quasar, 0957+561. If we observe the complete UV to optical spectrum of each component separated in time by the known time delay in this system, we can make a differential measurement of the reddening which does not depend on the intrinsic continuum shapes. This will yield a precise measurement of the dust-to-gas ratio in an absorption system at redshift z=1.3911 and will test whether the extinction laws of the Galaxy and Magellanic clouds are the same as those in extragalactic systems. We can also test our assumptions about how abundance measurements translate into dust- to-gas ratios by comparing these results with dust-to-gas ratios derived using high- reso lution spectra of both components obtained at Keck Observatory. The UV portion of these observations is required to discriminate between different extinction laws and clearly requires HST. Because the extinction in the optical is lower, very high accuracy in the relative flux calibration of the two objects is required. This accuracy is not attainable from the ground because of cross-contamination of the two spectra and contamination by the lensing galaxy. ------------------------------------------------------------------------------------ 8336( 5) - 04/05/00 09:33 - [ 2] Observations Description ------------------------ We will make identical observations of each quasar separated by the time delay. Component A of 0957+561 (given in the target list) will be observed in cycle 8 (2 orbits). Component B (10 1 20.9 +55 53 49) will be observed in cycle 9 (2 orbits) or in cycle 10 if component A is observed very late in cycle 8. We will use the STIS with both the near-UV MAMA and CCD. Target acquisition will consist of an ACQ of an offset object, followed by a move to the target and another ACQ. The offset target is required because it is possible, although unlikely, that the initial pointing would acquire the wrong target. Because the targets are 5" apart, the ACQ will work for a pointing after an offset. The offset star is the nearest star to the target in the guide star catalog. The magnitude, 14.6, is taken from that catalog, but no spectral type is available. We calculated the ACQ exposure time for the offset target by assuming that the spectral type is OV (yielding the longest time of a ny spectral type) and multiplying the output from the web based calculator (0.339s) by 4.1 to yield 1.4 seconds. We then calculated the count rate as if the offset star were type M6, which yields a count in the peak pixel of 20000, well short of saturation. For the ACQ observation of 0957+561A, we used converted g magnitude from Kundic et al. and spectral slopes measured by us as input to the web based calculator. The magnitude error are an estimate of the variablilty. Using V=17.1, we calculated an exposure time of 1.85 s, and have arbitrarily increased this to 10s for the proposal submission. We will use the three low resolution gratings to cover the entire spectrum from the Lyman limit at 2180 Angstrom to 1 micron. We will attain a signal-to-noise of better than 20 per resolution element over the entire spectrum (and better than 30 at most wavelengths) with exposures of 900s and 700s with G430L and G750L (the first orbit) and 2000s with G230L (the second orbit). These times are based on the rps out put with a SCHED 90% special requi rement. The signal-to-noise estimates were made with the WWW version of the STIS exposure calculator, using fluxes and spectral indices measured from our own data. For example, for component A, we have a flux of 1.7 * 10^-15 erg\cm^2\sec\Angstrom at 2200 Angstrom with Alpha=0.86 (f_Nu proportional nu^-Alpha) which gives approximately 480 cts/pixel at 2500 Angstrom, for a signal-to-noise per resolution element of slightly more than 30. The signal- to-noise is above 20 from the blue limit of the quasar spectrum (which is cut off by the Lyman limit near 2180 Angstrom) to near where the spectrum from the G430L starts. We will use the 0.2" slit rather than observing slitless to ensure that the spectra of the other component and G1 do not overlap the spectrum of interest regardless of roll angle. If the other objects fall on the slit, their spectra will be excluded in the reduction process. Because we have timing restrictions on the observations, we do not wish to have any orientation restrictions. A contemporaneous "fringe" flat field observations is required with each G750L observation. As described in chapter 7 of the STIS Instrument Handbook, this is needed because otherwise fringing would degrade the signal-to-noise at the red end of the spectrum below acceptable values. In practice, we will perform simultaneous fits to the data. However, we can assess the accuracy of our results by looking at the data we will obtain with G750L. In this portio n of the spectrum, the extinction law is A(Lambda) is proportional to tau_Lambda is proportional to 1/Lambda, and this leads to f_Lambda(A)/f_Lambd(B)= C-(1-N_B/N_A)*(N_A/10^21) * 4400 Angstrom * k * (1/Lambda). So a measurement of k is equivalent to measuring the slope of the straight line (red) portion of the curve in figure 1. For approximately 500 independent data points, with a signal-to-noise of 20 in each of the original spectra, we can derive a 3 sigma measurement for k=0.06. Note that any systematic errors such as wavelength dependent flux calibration uncertainties will divide out, provided they remain constant. Any reddening from galaxy G1 should also be small, because it is an elliptical galaxy. It will be additionally minimized because at z=0.36, we will be observing at much longer wavelengths (lower extinction) in the galaxy G1 frame than in the absorber frame. If k is not the same along both lines-of-sight, then we have (f_Lambda^A/f_Lambda^B)= C-(1-( k_B*N_B)/(k_A*N_A)*(N_A/10^21)*440 0 Angstrom * k_A * (1/Lambda). If k_B=(1+epsilon)*k_A (where epsilon is not restricted to small numbers), then for values of N_B and N_A given in the "Scientific Justification", if we compute k assuming that the values of k_B and k_A are equal, then (k_actual/k_computed) = 1/(1-0.73*epsilon). Thus, a 10% difference between k_A and K_B would lead to an 8% error in k. If epsilon is large, then the error in the calculation of k becomes significant. However, as described in the scientific justification, differences in k are likely to be correlated with differences in metal abundance (Vladilo 1998). These are likely to be small considering the small difference between the impact parameters of the two light paths, and in any case the metal abundance differences will be measured using high resolution optical observations (subject to any effects of dust depletion on metal abundances, of course). Real Time Justification ----------------------- Because the flux of this object is variable, and the shape of quasar spectra is known to vary with brightness changes, intrinsic variability of the object renders our measurement impossible if the observations are made at the same time in our frame, as described in the scientific justification. In order to compare the spectra of the two lensed components at the same time in the source frame, the observation must be separated in time by the time delay introduced by the gravitational lensing. The best current estimate for the time delay is approximately 417 +/- 3 days. Although the first observation in this series can be made at any time it is feasible, the second observation must come approximately 417 days later. There is one large window per 12 month period in which the first object can be observed and have the 417 day delay also fall into an observation window. Hence obtaining observations with the appropriate separation (+/- a few days) should not present a signif icant scheduling problem. In order to avoid any possiblility of variability of the object between the CCD and MAMA observations, they should ideally be made in the same visit. This would also ensure that the identical pointing be used for each observation. As the rules require the MAMA and CCD observations be conducted during separate visits, we have requested that both visits occur within one day of each other. One of the goals of this proposal is the comparison of the determination of the dust-to-gas ratio by observations of the two components with that determined from the more conventional measurements of weak absorption lines. Published spectra (Pettini et al. 1998) do not detect these lines. However, they should be measurable in higher resolution, higher signal-to-noise spectra obtainable with the HIRES spectrograph on the Keck I telescope. Another group has already obtained such observations, and we plan to use their results, when published, if they are adequate to our purposes (if this proposal is a ccepted, we can also discuss possi ble collaborations). If not suitable, we will obtain our own observations. A proposal we submitted to Keck observatory received very favorable comments, but was not accepted when it became known that another group already had data covering the wavelength range of interest. Thus, we believe that should the results of that observation prove unsuitable for our needs, a new proposal would likely be looked on favorably. Calibration Justification ------------------------- Our observations with G750L will extend beyond 7500 Angstroms. According to the STIS Instrument Handbook (chapter 7), the signal-to-noise of observations in that spectral region is limited by fringing. Thus, we require a contemporaneous "fringe flat" for each observation with G750L. Our limited experience (one red STIS observation) is in agreement with this. Re-calibration with the "fringe flat" was a great improvement over the pipline reduction. Additional Comments ------------------- ------------------------------------------------------------------------------------ 8336( 5) - 04/05/00 09:33 - [ 3] Data Distribution Paper Products: YES Media: 8MM Blocking Factor: 10 Ship To: PI_Address Ship Via: Email: ------------------------------------------------------------------------------------ 8336( 5) - 04/05/00 09:33 - [ 4] TARGET LIST Fixed Targets ------------------------------------------------------------------------------------------------------------------------------------ Tar| Target | Target | Target |Coord | Radial | Flux data No | Name | Description | Position |Eqnx | Vel. | ------------------------------------------------------------------------------------------------------------------------------------ 1 0957+561A GALAXY, QUASAR, EXT- RA=10H 1M 20.74S +/- 0.13", J2000 Z = V = 17.1+/-0.2 MEDIUM, ABSORPTION LINE DEC=55D 53' 55.1" +/- 0.13", 1.413 F(2600)=1.4+/-0.2e-15 SYSTEM - EXTRAGALACTIC, PLATE-ID=01RP DAMPED LYMAN ALPHA 2 0957+561B GALAXY, QUASAR, EXT- RA=10H 1M 20.89S +/- 0.13", J2000 Z = V = 16.9+/-0.2 MEDIUM, ABSORPTION LINE DEC=55D 53' 49.2" +/- 0.13", 1.413 F(2600)=2+/-0.2e-15 SYSTEM - EXTRAGALACTIC, PLATE-ID=01RP DAMPED LYMAN ALPHA Comments: This target will be observed 417 +/- 3 days after the observations of 0957+561A and is listed here for reference only. 3 0957+561- STAR RA=10H 1M 27.78S +/- 0.33", J2000 V = 14.6+/-0.5 OFFSET DEC=55D 53' 28.0" +/- 0.33", PLATE-ID=01RP Comments: The closest object in the guide star catalog to the program object. Actual error in magnitude should be lower than quoted here. 8336( 5) - 04/05/00 09:33 - [ 5] Visit: 01 Visit Priority: Visit Requirements: SCHED 90% BETWEEN 18-Feb-1999 AND 11-May-1999:12:00:00.0 On Hold Comments: Additional Comments: This observation is to be done 417 +/- 3 day prior to an observation of 0957+561B (5" south). When scheduling this observation, it should be tested whether the subsequent observation will also be doable on the appropriate date. The BETWEEN special requirement ensures this to the extent it could be checked with RPS2 (it was calculated with both targets in a trial proposal). If the first observing opportunity is missed, RPS2 shows another window from 5-September-1999 to 13-October-1999. I believe that the end date on that observing window is due to an artificial restriction on the last date upon which RPS2 could schedule the second observation. The true observing window is a yearly window running approximately from 5-September of a given year to 11-May of the next year. This is based on the Sun constraint (which is the only real constraint) and the 417 day lag between the two observations. Exposures ------------------------------------------------------------------------------------------------------------------------------------ Exposure| Target |Instr | Oper. | Aper |Spectral|Central| Optional |Num| Time | Special Number | Name |Config| Mode |or FOV |Element |Waveln.| Parameters |Exp| | Requirements ------------------------------------------------------------------------------------------------------------------------------------ 10 0957+561-OF STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 1.4 S ONBOARD ACQ FOR 20 FSET CD P ------------------------------------------------------------------------------------------------------------------------------------ 20 0957+561A STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 10.0 S ONBOARD ACQ FOR 30, 40 CD P ------------------------------------------------------------------------------------------------------------------------------------ 30 0957+561A STIS/C ACCUM 52X0.2 G430L 4300 1 750 S CD ------------------------------------------------------------------------------------------------------------------------------------ 40 0957+561A STIS/C ACCUM 52X0.2 G750L 7751 1 550 S CD ------------------------------------------------------------------------------------------------------------------------------------ 50 CCDFLAT STIS/C ACCUM 52X0.2 G750L 7751 2 DEF CD ------------------------------------------------------------------------------------------------------------------------------------ 8336( 5) - 04/05/00 09:33 - [ 6] Visit: 02 Visit Priority: Visit Requirements: GROUP 01, 02 WITHIN 1 D SCHED 90% On Hold Comments: Additional Comments: Exposures ------------------------------------------------------------------------------------------------------------------------------------ Exposure| Target |Instr | Oper. | Aper |Spectral|Central| Optional |Num| Time | Special Number | Name |Config| Mode |or FOV |Element |Waveln.| Parameters |Exp| | Requirements ------------------------------------------------------------------------------------------------------------------------------------ 10 0957+561-OF STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 1.4 S ONBOARD ACQ FOR 20 FSET CD P ------------------------------------------------------------------------------------------------------------------------------------ 20 0957+561A STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 10.0 S ONBOARD ACQ FOR 30 CD P ------------------------------------------------------------------------------------------------------------------------------------ 30 0957+561A STIS/N ACCUM 52X0.2 G230L 2376 1 2200 S UV-MAM A ------------------------------------------------------------------------------------------------------------------------------------ 8336( 5) - 04/05/00 09:33 - [ 7] Visit: 03 Visit Priority: Visit Requirements: SCHED 90% AFTER 01 BY 414 D TO 420 D GROUP 3-4 WITHIN 1 D On Hold Comments: Additional Comments: This observation is to be done 417 +/- 3 day prior to an observation of 0957+561B (5" south). When scheduling this observation, it should be tested whether the subsequent observation will also be doable on the appropriate date. The BETWEEN special requirement ensures this to the extent it could be checked with RPS2 (it was calculated with both targets in a trial proposal). If the first observing opportunity is missed, RPS2 shows another window from 5-September-1999 to 13-October-1999. I believe that the end date on that observing window is due to an artificial restriction on the last date upon which RPS2 could schedule the second observation. The true observing window is a yearly window running approximately from 5-September of a given year to 11-May of the next year. This is based on the Sun constraint (which is the only real constraint) and the 417 day lag between the two observations. Exposures ------------------------------------------------------------------------------------------------------------------------------------ Exposure| Target |Instr | Oper. | Aper |Spectral|Central| Optional |Num| Time | Special Number | Name |Config| Mode |or FOV |Element |Waveln.| Parameters |Exp| | Requirements ------------------------------------------------------------------------------------------------------------------------------------ 10 0957+561-OF STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 1.4 S FSET CD P ------------------------------------------------------------------------------------------------------------------------------------ 20 0957+561B STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 10.0 S CD P ------------------------------------------------------------------------------------------------------------------------------------ 30 0957+561B STIS/C ACCUM 52X0.2 G430L 4300 1 750 S CD ------------------------------------------------------------------------------------------------------------------------------------ 40 0957+561B STIS/C ACCUM 52X0.2 G750L 7751 1 550 S CD ------------------------------------------------------------------------------------------------------------------------------------ 50 CCDFLAT STIS/C ACCUM 52X0.2 G750L 7751 2 DEF CD ------------------------------------------------------------------------------------------------------------------------------------ 8336( 5) - 04/05/00 09:33 - [ 8] Visit: 04 Visit Priority: Visit Requirements: SCHED 90% On Hold Comments: Additional Comments: Exposures ------------------------------------------------------------------------------------------------------------------------------------ Exposure| Target |Instr | Oper. | Aper |Spectral|Central| Optional |Num| Time | Special Number | Name |Config| Mode |or FOV |Element |Waveln.| Parameters |Exp| | Requirements ------------------------------------------------------------------------------------------------------------------------------------ 10 0957+561-OF STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 1.4 S FSET CD P ------------------------------------------------------------------------------------------------------------------------------------ 20 0957+561B STIS/C ACQ F28X50L MIRROR ACQTYPE=POINT 1 10.0 S CD P ------------------------------------------------------------------------------------------------------------------------------------ 30 0957+561B STIS/N ACCUM 52X0.2 G230L 2376 1 2200 S UV-MAM A ------------------------------------------------------------------------------------------------------------------------------------ 8336( 5) - 04/05/00 09:33 - [ 9] Summary Form for Proposal 8336 Item Used in this proposal ------------------------------------------------------------------------------------------------------------------------------------ Apertures F28X50LP 52X0.2 ------------------------------------------------------------------------------------------------------------------------------------ Configurations STIS/CCD STIS/NUV-MAMA ------------------------------------------------------------------------------------------------------------------------------------ Opmodes ACQ ACCUM ------------------------------------------------------------------------------------------------------------------------------------ Optional Parameters ACQTYPE=POINT ------------------------------------------------------------------------------------------------------------------------------------ Proposal Category GO ------------------------------------------------------------------------------------------------------------------------------------ Special Requirements SCHED 90% BETWEEN 18-Feb-1999 AND 11-May-1999 12 00 00.0 ONBOARD ACQ FOR 20 ONBOARD ACQ FOR 30, 40 GROUP 01, 02 WITHIN 1 D SCHED 90% ONBOARD ACQ FOR 30 SCHED 90% AFTER 01 BY 414 D TO 420 D GROUP 3-4 WITHIN 1 D SCHED 90% ------------------------------------------------------------------------------------------------------------------------------------ Spectral Elements MIRROR G430L G750L G230L ------------------------------------------------------------------------------------------------------------------------------------ Target Names 0957+561-OFFSET 0957+561A CCDFLAT 0957+561B ------------------------------------------------------------------------------------------------------------------------------------ Wavelengths 4300 7751 2376 ------------------------------------------------------------------------------------------------------------------------------------